You say that on a very fine but still night, water is cooled less rapidly than the earth:  under such circumstances the bottom of the pond cools more rapidly than the surface, the plants become colder—­in fact, some degrees below the freezing-point, &c. &c.

I submit that such reasons are inadmissible, for there would be an immediate upward current, which, as water is such an excellent conductor of heat, would immediately equalize the temperature of all the water above 40 degrees Fahrenheit, and stratified (if I may use the expression) above the water of this temperature there would be another layer of water of equal but gradually decreasing temperature until it fell below 32 degrees Fahrenheit.

The explanation I offer is this.  It is well known that if water is kept perfectly still it may be cooled down considerably, or at least some degrees below 32 degrees, without freezing; but the moment it is shaken a portion of it is converted into a spongy, porous ice, and the temperature rises to 32 degrees.

What may be the case in the rivers of the South of England I do not know, but in the rapid streams of the North this process may be seen on a very extensive scale in severe frosts.  The water in the still pools (before they are frozen over) is cooled down to below 32 degrees, and so soon as this cooled water reaches the next stream, precipitation (if I may so call it) takes place, and the spongy ice lays hold of every projecting pebble, which serves as a nucleus in the same way as threads and bits of stick serve in the crystallization of salts.  After a severe frost, when followed by bright sunshine the next morning, I have seen thousands of these bits of spongy ice rising from the stones to which they had been attached to the surface of the water.  I have seen after long-continued frost the course of a stream completely altered by this bottom-ice (as it is called here), and I have also seen a weir with a wall of ice on it three feet high (raised in a single night) by the same cause.  Now apply this to the bottom-ice in ponds (which however I must confess I never saw).  The night being calm, the water gets cool below 32 degrees, but then a breeze springing up the water becomes agitated, precipitation takes place, and the plants serving as nuclei become immediately clothed with this spongy ice, and the sun shining next morning imparts so much warmth to the plants that the ice thaws which is in contact with them, and rises to the surface.  Of course if the sun does not shine next morning, and the frost continues, the plants may be clothed with ice for a long time.

To the foregoing the Editor of the “Agricultural Gazette” replied as follows:—­

We cannot admit the soundness of our correspondent’s explanation of the formation of bottom-ice or ground gore.  We are well acquainted with the statements of Arago and other writers as to the cause of this curious phenomenon, and after a careful consideration of the subject believe that it is due to radiation and not to any other cause.  Bottom-ice has been observed in ponds on perfectly still nights when there was no breeze to agitate the surface of the water.